Skip to main navigation Skip to search Skip to main content

Locating Image Background By Monotonic Tree

  • SUNY Buffalo

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

In this paper, we discuss an approach to locating image background, which is based on monotonic tree, a derivation of contour tree for discrete data. We first make two assumptions about image background. As compared with the foreground shapes, the background regions are (1) smoother in intensity; and (2) peripheral in location. In an input image, the harshness of the foreground is characterized by intensity peaks and valleys, which are modeled in monotonic tree as small textons whose altitudes are high. Based on this, we can find the texton pattern for foreground objects. Then the foreground region, and thus the background region are rendered. Experiments show that our scheme can effectively locate background in images which satisfy the two assumptions above.

Original languageEnglish
Title of host publicationProceedings of the 6th Joint Conference on Information Sciences, JCIS 2002
EditorsJ.H. Caulfield, S.H. Chen, H.D. Cheng, R. Duro, J.H. Caufield, S.H. Chen, H.D. Cheng, R. Duro, V. Honavar
Pages879-884
Number of pages6
StatePublished - 2002
EventProceedings of the 6th Joint Conference on Information Sciences, JCIS 2002 - Research Triange Park, NC, United States
Duration: Mar 8 2002Mar 13 2002

Publication series

NameProceedings of the Joint Conference on Information Sciences
Volume6

Conference

ConferenceProceedings of the 6th Joint Conference on Information Sciences, JCIS 2002
Country/TerritoryUnited States
CityResearch Triange Park, NC
Period03/8/0203/13/02

Fingerprint

Dive into the research topics of 'Locating Image Background By Monotonic Tree'. Together they form a unique fingerprint.

Cite this